POsted on behalf of Doug Pillars

I am trying to find a “best Practice” matrix for over all maintenance performances?

80% preventive maintenance, 100% of all work through a CMMS system and so on.

I would appreciate it f some one could e-mail me the ruler that we are all measured by. I had this but seem to have lost it in a computer migration incident.

Thank you in advance

Doug Pillars
Maintenance/Reliability Engineer
Nissan

Doug -

You can find one such matrix at MaintenanceBenchmarking.com from John Mitchell's Book "Physical Asset Management" http://www.maintenancebenchmarking.com/best_practice_maintenance.htm.

You can also find a other benchmarking projects http://www.maintenancebenchmarking.com/

Good luck,
Terry O'Hanlon, CMRP

Doug,

I understand your thinking on the need for 100% of jobs through the CMMS, but 80% Preventive?

What is holy about Preventive Maintenance? What is unacceptable about Breakdown Maintenance when the consequences of failure are trivial or zero? If you have suffient installed spare capacity or carry intermediate buffer storage, perhaps PM will be hard to justify. If you can predict the degradation mechanism reasonably accurately, Condition-basedc PdM can be very effective.

My point is that the ratio of Preventive to Total work cannot be decided till you know what the consequences of failure are, and that can only be determined by the configuration and local situation in your Plant. We have to ensure the lowest total risk; that is not always the same thing as maximum PM.

V.Narayan.

I would compare reactive workoders x planned (PM, PdM others)

The goal Reactive workorders > 20%

Reactive workorders are jobs that typically done without advanced planning (less then a week). Planning must be done on the fly, you will probably have unexpected overtime.

Imagine scenario1

Burner fuel oil pressure is decreasing in time. Minimum pressure is 150 psi, below that no atomizing of fuel is possible. Normal is 300- 350 psi.
Somewhere at 250 psi, someone alerts maintenance, shut down is scheduled, craft and tools are prepared. Probable causes are listed, is it the burner tip? is it the fuel pump? or is it the controller?, preliminary checks are done and conditions are monitored. (planned)

Scenario 2
Maintenance comes in Monday morning and hears emergency shut down in the weekend, the furnace is full of oil !! (Reactive!!)

This job definitely does not fit in the Preventive Maintenance Category, but it happens all the time.

Sorry

The goal Reactive workorders < 20%

quote:

---

Originally posted by svanels:
Sorry

The goal Reactive workorders < 20%

---

Your example is of a high consequence failure, on which I have no problem with a PM or PdM activity.

However, there are many failures with little or no effect on Safety or Production. Typically, if you have installed spare equipment, the duty eqpt. failure should have no effect. Other than minimum lub. and very basic condition monitoring, I would argue that NO other maintenance is justified,and run-to-failure is acceptable, unless the breakdown also causes serious internal machine damage.

So where does this magic threshold of 20% come from? Why not 10, 30 or 50%? Each situation is different, and should be treated on merit. Any other method is less tthan optimal.

V.Narayan.

Vee I also heard this myth of 80% PPM (periodic preventive maintenance) but never came across something written about it.

I only encountered the 80 - 20 rule in planned x reactive workorders in "Developing Performance Indicators for Managing Maintenance" by Terry Wireman.

The run to failure method depends on the type of equipment, and his production function. The run to failure method is suitable for the wiper blades of car, you wouldn't change the blades every two 6 months.
In a production environment, things are different. We have installed spare equipment and installed critical spare equipment.
Fortunately they are fixed and it is bare minimum (2).

If the spare equipment is mobile and you have more then 2, you have no spare. Because Operations will notify that there are problems after they canibalized everything, especially in a 24 hours operation.

I would be very comfortable with 0% reactive.
80% Preventive Maintenance? Maybe it would fit for certain equipment, but not as maintenance strategy for an entire plant.

Dear ALL

As you all aware, many people are aiming for 80%PM & 20%CM. I wonder once this magic figure has been achieved, what would be the total cost of all the PM and CM works and the plant uptime%?

I think many people are aiming to achieve optimum uptime such as 97% but will the total costfigure remain within <3% of CPRV and <10-15% of total manufacturing cost?

Any actual results being published which reinforced the above thought which appears to be the right direction to go. I just want to proofs that the Mitchell's benchmarks are realistic for implementation. So I hope to see more actual results versus the Mitchell's bechmarks. TQ

Dear Josh,

I can understand aiming for a 97% availability, also that of 80% of work being Planned, but I do not know why 80% of work should be Preventive. In fact Breakdown maintenance is a perfectly acceptable strategy under certain circumstances, and can be as high as 50% of the total in some cases. When you do an RCM, it is not unusual to find that for many failure modes, that is the most cost-effective strategy. It all depends on the configuration, how many installed spare units are available, intermediate storage capacity etc. So while we can strive to Plan 80%+ of our maintenance work, it can include the planning of breakdown work too. Obviously, scheduling breakdown work is more difficult, since the timing of breakdowns is uncertain, but there is no reason why should not Plan (i.e., think through all the steps we have to follow to execute the work, evaluate risks, estimate resources, identify procedures, drawings etc.).

It seems fashionable (but not cost-effective) to prevent all failures. This is seen as pro-active, so it must be good. I do not support this view. It takes equipment out of service for the wrong reasons, costs money and does not add value.

When the consequences matter, then we can justify preventive work or condition monitoring. In a manufacturing Plant with single line production paths and little or no intermediate storage, failure consequences are usually high. Here a very high proportion of Preventive work is justified. In other situations, this varies. Regards.

V.Narayan.

I agree with the views presented. My plant is a single production line. Thus the general aim is to achieve 80%PM works by cost (of manhours, materials & contracts) and the remaining 20% for CM & BM works.

My reservation is that once we achieve this figure, will the plant uptime be at 97% and the total cost of all works (80%PM & 20%CM) is considered cost-effective?

Have you seen any actual results published that proves the above scenario is worth-while to accomplish & perservere? TQ

Dear Josh,

I am afraid not; you need to do a couple of other things as well before you can expect good rtesults.

The first is compliance. You have to do what you planned to do, on time. In other words, if you said you will do it on a given date, you have to do it within a tolerance band around this date.

The second is work quality. If the work done does not meet the required standards, e.g., in terms of cleanliness, alignment, balancing, clearances, tightness etc., the paper plan is just that, no more.

V.Narayan.

As per this maintenance KPI scorecard, it appears that we should aim for a high proportion of PM works over all maint works that is >85% whihc coincides with >97% availability:
http://www.maintenancebenchmarking.com/best_practice_maintenance.htm.

Josh,

As we have been over in a few other forums recently. The relation between PM (by which I believe you are talking about routine maitnenance and not just preventive maintenance tasks) and availability is nowhere near as simple as the quote you place here would have you think.

quote:

As per this maintenance KPI scorecard, it appears that we should aim for a high proportion of PM works over all maint works that is >85% whihc coincides with >97% availability

In fact there is a tendancy around these days to oversimplify maitnenance in general. This is a trend that defies logic and is not safe in my belief.

The level of routine maintenance, (whether preventive, predictive or detective), is determined by the configuration of the plant, the level of risk which an organization is willing to tolerate, and the consequences of failure of each failure mode.

A plant where there is a lot of stand by equipment may wind up with a maintenance routine that accepts a large percentage of "planned corrective works". (Or run-to-failure) as this may be the more economically effective means of managing the plant.

Similarly a plant where there is a lack of stand by equipment may find that there is a need for a high portion of routine maintenance. (Based on other factors of course)

In fact, over maintenance is one of the key causes of reduced reliability. So the notion of merely more routine maitnenance equals lower costs and higher availability is surprising in the 21st century.

Cheers

Daryl Mather
www.strategic-advantages.com

Josh, the link is broken, but in the context of the document
Planned Maintenance (PM) versus Reactive Maintenance

Preventive or Preventative Maintenance is also PM, and I think that is the common abreviation used by maintenance professionals, or am I wrong?

Yes you are right. Planned maintenance includes planned preventive & predictive maintenance plans plus planned inspection plans. These can either be time based or conditioned based. These planned works can be carried out during operations or turnaround.

Other works which are not planned are reactive (waiting for operations personnel or equipment to notify problems to maintenane dept)

So the above is what I mean the % ratio of PM over RM. TQ

It appears that there is a view saying that amount of planned maintenance (PM) relative to reactive amintenance (RM) depends on plant configuration such as having standby equipment.

My current understanding is that standby equipment are only affordable for small equipment like pumps & fans but for critical & normally complex large equipment cannot affordable to be parallel eg fired heaters, boilers, heat exchangers, columns, compressor & turbunes whose trips will surely results in production/HSElosses.

Is there a way to quantify the whole facility reliability based on its configuration and from this, we deduce how much planned works we should have? TQ

I deliberately don't want to use Planned Maintenance, because the original post was about 80 - 20% preventive maintenance

The ratio Planned Workorders x Reactive Workorders, In my view doesn't depend much on plant configuration.

It tells you more about how maintenance is executed. 80% reactive means, high costs, poor work, lots of overtime, stress and probably unskilled workers, supervisors and incompetent management.

Dear Svanels,

You have hit the nail on its head! Going for 80%+ Planned is good, but in my view, Preventive maintenance can be high or low depending on the risks involved.

I hear a lot about reactive maintenance in a somewhat derogatory way. In my view reactive maintenance is perfectly acceptable, where the risks are low. Reactive maintenance can and in some cases should be planned, i.e., thought through in advance based on expected scenarios. The work steps and sequence can be identified, spares and materials listed, drawings and procedures collected, Job Safety Analysis done if relevant, and other risks identified well in advance. Because the timing of such work is not known, they cant be scheduled in advance, but nothing stops us from planning them.

There seems to ba a lot of confusion about terminology. Regards,

V.Narayan.

Josh, one of the first things to be defined is the list of critical equipment.
We run a small refinery and out of the list of 200+ equipment, we defined 20 pieces as critical equipment
The list consists of 2 steamboilers, 2 diesel driven generators, plant air compressors, bottoms pumps and fire water pumps. The list has been elaborated by maintenance + operations and sanctioned by management.

The philosophy behind this list is that loss of functionality of these systems will jeopardize directly production and/or introduce hazards to humans and installations.

We have heaters, but you cannot take them off-line to do service. One is decoked once a year the other runs continously for 4 years. They are critical for the process, but not from a day to day maintenance point of view. The same applies to heat exchangers, columns and vessels. These items are serviced during a turnaround.

But wait, there is also the non-official critical list. The central airco of the director's office is on it. When that stops the phones keep on ringing...